Treatment with nicotinic acid has been found to be highly effective for lowering levels of total cholesterol, low-density lipoprotein cholesterol, triglycerides, and perhaps lipoprotein Lp (a), raising levels of high-density lipoprotein cholesterol, and reducing mortality from coronary heart disease as well as total mortality. An important drawback of nicotinic acid therapy is that there has not yet been a systematic appraisal of the side effects (nausea, blurred vision, cutaneous flushing) and medical complications (diabetes, peptic ulcer, gouty arthritis, liver dysfunction) that have been associated with its use. Thus, clinicians lack the conviction that nicotinic acid can be safely and effectively managed. We propose to perform an indepth appraisal of the management of nicotinic acid. Data from the Coronary Drug Project -- a 6-year, double-blind, randomized, controlled clinical trial of nicotinic acid use in 1119 males -- have been obtained and will be the major source of information. Subjects enrolled in this trial were assessed every 4 months; measures of liver function, glucose tolerance, and levels of blood lipids were obtained, and a thorough physical exam and assessment of treatment side effects conducted. The primary objective of this proposal is to identify and characterize key variables associated with the development of side effects and medical complications, and use this information to develop a clinical management model. Patterns of change for each of the biochemical variables of interest (e.g., blood levels of aspartate amino transferase, alkaline phosphatase, glucose, uric acid, total cholesterol, triglycerides) will be determined and comparisons made between the placebo and treatment cohorts to characterize the toxic response to nicotinic acid. Then, a search for variables (e.g., age, body, weight, other medical conditions) that predict abnormal changes in the biochemical parameters will be conducted. Finally, logistic regression methods will be used to 1) develop models that predict which patients have a susceptibility to complications arising from nicotinic acid use, so treatment can be avoided in those, and 2) determine which variables are early predictors of an impending toxic response, so treatment can be appropriately modified. Nicotinic acid is presently sixteen times less expensive than lovastatin, a dyslipidemia drug with comparable efficacy. With its demonstrated record of efficacy and cost-effectiveness, removing the last obstacle for widespread use of nicotinic acid in appropriate patients -- the physician's lack of confidence in effective management -- would provide an important tool for the treatment of dyslipidemia, and consequently have a beneficial impact upon rates of heart disease mortality.